The pilot had recent experience operating from Noganosh Lake, and, prior to departing from the main base, he had discussed the trip with a senior pilot. It is not known why the pilot did not taxi south into Turner's Bay to utilize all of the available water for the take-off; there was no known operational or other requirement to expedite the departure from Noganosh Lake. On the Friday prior to the accident, the pilot departed from approximately the same position on the lake as on the accident flight. Although the take-off and climb-out during Friday's flight was uneventful, the conditions under which the take-off was performed were different. It is possible that on the occurrence flight, the pilot operated the aircraft in a similar manner to the departure of the previous Friday. He may have applied erroneously the experience gained on that day, and not fully comprehended the effect of the higher take-off weight and the different wind and water conditions on the take-off distance required. Pilot take-off technique may have contributed to the length of the take-off run. It is not known why the pilot did not reject the take-off; however, several possibilities exist. The pilot may not have chosen a reject take-off point prior to beginning the take-off, or he may have chosen a reject take-off point farther along the lake from where the aircraft became airborne, or he may have formulated another plan to follow during the climb-out. In any event, he did not abort the take-off. Once he entered the narrows toward the north end of the lake, it is possible that he encountered descending air on the leeward side of the island and/or a tail wind resulting from a funnelling effect through the narrows. Either or both of these factors would have had a detrimental effect on aircraft performance at a critical time of the flight. If the pilot planned a later reject take-off point or had an alternative plan to follow for the climb-out, an unexpected downdraft and/or tail wind condition would have compelled him to deviate from or modify his plan. In considering the lift-off point, the terrain ahead of the aircraft after lift-off, and the aircraft climb performance, the pilot likely had three possible options. First, he could set the aircraft back down on the water and attempt to slow or stop the aircraft before reaching the shoreline. Second, he could turn to the left and attempt to negotiate the passage to the northern section of the lake while airborne. Or third, he could turn the aircraft further to the left to negotiate the passage between the island and the shoreline while airborne. The pilot chose the latter option, possibly to take advantage of the anticipated improvement in aircraft climb performance from the prevailing winds. Also, the passage to the north portion of the lake is initially a narrower passage and this may have dissuaded him from taking that route. The continued turn to the left required a steep left turn, followed by a steep right turn and a second steep left turn to negotiate the passage. The fact that the pilot was able to successfully manoeuvre between the island and the shoreline suggests that there were no controllability problems with the aircraft. The steep bank angle required to execute the turns around the island would have precluded any appreciable altitude gain, thus preventing him from climbing above the terrain. When the aircraft emerged from the passage between the island and the shoreline, there was insufficient open space to climb above the rising terrain ahead. A skidding turn at impact was probably a result of excessive left rudder input, most likely made in an attempt to increase the rate of turn to avoid the collision with the trees.2.0 Analysis The pilot had recent experience operating from Noganosh Lake, and, prior to departing from the main base, he had discussed the trip with a senior pilot. It is not known why the pilot did not taxi south into Turner's Bay to utilize all of the available water for the take-off; there was no known operational or other requirement to expedite the departure from Noganosh Lake. On the Friday prior to the accident, the pilot departed from approximately the same position on the lake as on the accident flight. Although the take-off and climb-out during Friday's flight was uneventful, the conditions under which the take-off was performed were different. It is possible that on the occurrence flight, the pilot operated the aircraft in a similar manner to the departure of the previous Friday. He may have applied erroneously the experience gained on that day, and not fully comprehended the effect of the higher take-off weight and the different wind and water conditions on the take-off distance required. Pilot take-off technique may have contributed to the length of the take-off run. It is not known why the pilot did not reject the take-off; however, several possibilities exist. The pilot may not have chosen a reject take-off point prior to beginning the take-off, or he may have chosen a reject take-off point farther along the lake from where the aircraft became airborne, or he may have formulated another plan to follow during the climb-out. In any event, he did not abort the take-off. Once he entered the narrows toward the north end of the lake, it is possible that he encountered descending air on the leeward side of the island and/or a tail wind resulting from a funnelling effect through the narrows. Either or both of these factors would have had a detrimental effect on aircraft performance at a critical time of the flight. If the pilot planned a later reject take-off point or had an alternative plan to follow for the climb-out, an unexpected downdraft and/or tail wind condition would have compelled him to deviate from or modify his plan. In considering the lift-off point, the terrain ahead of the aircraft after lift-off, and the aircraft climb performance, the pilot likely had three possible options. First, he could set the aircraft back down on the water and attempt to slow or stop the aircraft before reaching the shoreline. Second, he could turn to the left and attempt to negotiate the passage to the northern section of the lake while airborne. Or third, he could turn the aircraft further to the left to negotiate the passage between the island and the shoreline while airborne. The pilot chose the latter option, possibly to take advantage of the anticipated improvement in aircraft climb performance from the prevailing winds. Also, the passage to the north portion of the lake is initially a narrower passage and this may have dissuaded him from taking that route. The continued turn to the left required a steep left turn, followed by a steep right turn and a second steep left turn to negotiate the passage. The fact that the pilot was able to successfully manoeuvre between the island and the shoreline suggests that there were no controllability problems with the aircraft. The steep bank angle required to execute the turns around the island would have precluded any appreciable altitude gain, thus preventing him from climbing above the terrain. When the aircraft emerged from the passage between the island and the shoreline, there was insufficient open space to climb above the rising terrain ahead. A skidding turn at impact was probably a result of excessive left rudder input, most likely made in an attempt to increase the rate of turn to avoid the collision with the trees. 3.0 Conclusions 3.1 Findings The pilot was certified and qualified for the flight in accordance with existing regulations. The aircraft was certified, equipped, and maintained in accordance with existing regulations and approved procedures. The weight and centre of gravity were within the prescribed limits. There was no evidence found of any airframe failure or system malfunction prior to or during the flight. The aircraft was not equipped with shoulder harnesses. The take-off distance on the occurrence flight was greater than the take-off distance on the previous departure from Noganosh Lake because of the aircraft's take-off weight and prevailing meteorological and water conditions. Pilot take-off technique may have contributed to the length of the take-off run. Approximately 3,600 feet of usable take-off distance remained behind the pilot when he began the take-off. The aircraft did not climb above the trees after take-off. It is possible that, after take-off, the aircraft encountered a downdraft, a tail wind, or both. 3.2 Causes The pilot began the take-off run from a position on the lake which did not allow sufficient distance for the take-off and climb-out. The Board has no aviation safety recommendations to issue at this time.4.0 Safety Action The Board has no aviation safety recommendations to issue at this time.